Speech dereverberation is one of the most difficult tasks in acoustic signal processing. Of the various problems involved in this task, this paper highlights “over-whitening,” which flattens the characteristics of recovered speech. This distortion sometimes happens when inverse filters are directly calculated from microphone signals. This paper reviews two studies related to this problem. The first study shows the possibility of compensating for such over-whitening to achieve precise speech-dereverberation. The second study presents a new approach for approximating the original speech by removing the effect of late reflections from observed reverberant speech.

抄録全体を表示

Although it is very important to conduct listening tests when constructing a practical sound field reproduction system based on wave field synthesis, listening tests are very expensive. A localization model of synthesized sound images that predicts the results of listening tests is proposed. This model reduces the costs of constructing a reproduction system because it makes it possible to omit the listening tests. The proposed model uses the precedence effect and predicts the direction of synthesized sound images based on the inter-aural time difference. A comparison of the results predicted by the proposed model and the localized results of listening tests shows that the model accurately predicts the localized results.

抄録全体を表示

Auditory artifacts due to switching head-related transfer functions (HRTFs) are investigated, using a software-implemented dynamic virtual auditory display (DVAD) developed by the authors. The DVAD responds to a listener's head rotation using a head-tracking device and switching HRTFs to present a highly realistic 3D virtual auditory space to the listener. The DVAD operates on Windows XP and does not require high-performance computers. A total system latency (TSL), which is the delay between head motion and the corresponding change of the ear input signal, is a significant factor of DVADs. The measured TSL of our DVAD is about 50ms, which is sufficient for practical applications and localization experiments. Another matter of concern is the auditory artifact in DVADs caused by switching HRTFs. Switching HRTFs gives rise to wave discontinuity of synthesized binaural signals, which can be perceived as click noises that degrade the quality of presented sound image. A subjective test and excitation patterns (EPNs) analysis using an auditory filter are performed with various source signals and HRTF spatial resolutions. The results of the subjective test reveal that click noise perception depends on the source signal and the HRTF spatial resolution. Furthermore, EPN analysis reveals that switching HRTFs significantly distorts the EPNs at the off signal frequencies. Such distortions, however, are masked perceptually by broad-bandwidth source signals, whereas they are not masked by narrow-bandwidth source signals, thereby making the click noise more detectable. A higher HRTF spatial resolution leads to smaller distortions. But, depending on the source signal, perceivable click noises still remain even with 0.5-degree spatial resolution, which is less than minimum audible angle (1 degree in front).

抄録全体を表示

In this paper, we propose a computationally efficient method of compensating temperature for the transaural stereo. The conventional method can be used to estimate the change in impulse responses caused by the fluctuation of temperature with high accuracy. However, the large amount of computation required makes real-time implementation difficult. Focusing on the fact that the amount of compensation depends on the length of the impulse response, we reduce the computation required by segmenting the impulse response. We segment the impulse responses in the time domain and estimate the effect of temperature fluctuation for each of the segments. By joining the processed segments, we obtain the compensated impulse response of the whole length. Experimental results show that the proposed method can reduce the computation required by a factor of nine without degradation of the accuracy.

抄録全体を表示

In this paper, we propose a two-microphone noise reduction method to deal with non-stationary interfering noises in multiple-noise-source environments in which the traditional two-microphone algorithms cannot function well. In the proposed algorithm, multiple interfering noise sources are regarded as one virtually integrated noise source in each subband, and the spectrum of the integrated noise is then estimated using its virtual direction of arrival. To do this, we suggest a direction finder for the integrated noise using only two microphones that performs well even in speech active periods. The noise spectrum estimate is further improved by integrating a single-channel noise estimation approach and then subtracted from that of the noisy signal, finally enhancing the desired target signal. The performance of the proposed algorithm is evaluated and compared with the traditional algorithms in various conditions. Experimental results demonstrate that the proposed algorithm outperforms the traditional algorithms in various conditions in terms of objective and subjective speech quality measures.

抄録全体を表示

We propose a new robust frequency domain acoustic echo cancellation filter that employs a normalized residual echo enhancement. By interpreting the conventional robust step-size control approaches as a statistical-model-based residual echo enhancement problem, the optimal step-size introduced in the most of conventional approaches is regarded as optimal only on the assumption that both the residual echo and the outlier in the error output signal are described by Gaussian distributions. However, the Gaussian-Gaussian mixture assumption does not always hold well, especially when both the residual echo and the outlier are speech signals (known as a double-talk situation). The proposed filtering scheme is based on the Gaussian-Laplacian mixture assumption for the signals normalized by the reference input signal amplitude. By comparing the performances of the proposed and conventional approaches through the simulations, we show that the Gaussian-Laplacian mixture assumption for the normalized signals can provide a better control scheme for the acoustic echo cancellation.

抄録全体を表示

We propose a new algorithm for the blind source separation (BSS) approach in which independent component analysis (ICA) and frequency subband beamforming interpolation are combined. The slow convergence of the optimization of the separation filters is a problem in ICA. Our approach to resolving this problem is based on the relationship between ICA and null beamforming (NBF). The proposed method consists of the following three parts: (I) a frequency subband selector part for learning ICA, (II) a frequency domain ICA part with direction-of-arrivals (DOA) estimation of sound sources, and (III) an interpolation part in which null beamforming constructed with the estimated DOA is used. The results of the signal separation experiments under a reverberant condition reveal that the convergence speed is superior to that of the conventional ICA-based BSS methods.

抄録全体を表示

A robust multichannel sound reproduction system that utilizes the relationship, between the width of the actual control area and the control frequency of the control points is proposed. The reproduction accuracy of a conventional sound reproduction system is reduced by room environment variations when fixed inverse filter coefficients are used. This tendency becomes more significant when control points are arranged more closely. To resolve this problem, the frequency control band at every control point is switched to avoid degrading the reproduced sound in low frequencies, so the pass band range of the control points at both ears is only high-range. That of the other control points is the entire control range. Numerical simulation with real environmental data showed that improvement of the reproduction accuracy is about 6.1dB on average, even with a temperature fluctuation of 5°C as an environmental variation in the listening room.

抄録全体を表示

Human interest in pictures dates back to 14,000 BC. Pictures can be drawn by hand or imaged by optical means. Over time pictures have changed from being rare and unique to ubiquitous and common. They have changed from treasures to transients. This paper summarizes many picture technologies, and discusses their dynamic range, their color and tone scale rendering. This paper discusses the interactions between advances in technology and the interests of its users over time. It is the combination of both technology and society's usage that has shaped imaging since its beginning and continues to do so.

抄録全体を表示

In this paper, the impact of display on quality assessment is addressed. Subjective quality assessment experiments have been performed on both LCD and CRT displays. Two sets of still images and two sets of moving pictures have been assessed using either an ACR or a SAMVIQ protocol. Altogether, eight experiments have been led. Results are presented and discussed, some differences are pointed out. Concerning moving pictures, these differences seem to be mainly due to LCD moving artefacts such as motion blur. LCD motion blur has been measured objectively and with psycho-physics experiments. A motion-blur metric based on the temporal characteristics of LCD can be defined. A prediction model have been then designed which predict the differences of perceived quality between CRT and LCD. This motion-blur-based model enables the estimation of perceived quality on LCD with respect to the perceived quality on CRT. Technical solutions to LCD motion blur can thus be evaluated on natural contents by this mean.

抄録全体を表示

This paper proposes a new color halftone image quality assessment method based upon the color structural similarity measure with considering the human visual characteristics. To include the color visual characteristics, we carry out the color filtering for each luminance, redgreen, and blue-yellow channels. Then, we apply the color structural similarity measure to the color filtered images, which are the reference image and the halftoned image, to evaluate the localized structural difference. By considering those characteristics, in this paper, the assessment of the color halftone images can be realized. We apply the proposed measure to the various kinds of color halftone images and confirm that the proposed measure can give reasonable results compared with the results by subjective evaluation.

抄録全体を表示

The TFT-LCD image has non-uniform brightness that is the major difficulty of finding the visible defect called Mura in the field. To facilitate Mura detection, background signal shading should level off and Mura signal must be amplified. In this paper, Mura signal amplification and background signal flattening method is proposed based on human visual system (HVS). The proposed DC normalized contrast sensitivity function (CSF) is used for the Mura signal amplification and polynomial regression (PR) is used to level off the background signal. In the enhanced image, trimodal thresholding segmentation technique is used for finding Dark and White Mura at the same time. To select reliable defect, falsely detected invisible region is eliminated based on Weber's Law. By the experimental results of artificially generated 1-d signal and TFT-LCD image, proposed algorithm has novel enhancement results and can be applied to real automated inspection system.

抄録全体を表示

In this paper, we show that motion sharpening phenomenon can be explained as a form of visual masking for a special case where a video sequence is composed of alternate frames with different level of sharpness. A frame of higher sharpness behaves to mask the ambiguity of a subsequent frame of lower sharpness and hence preserves the perceptive quality of the whole sequence. Borrowing the mechanism for visual masking, we formulated a quantitative model for deriving the minimum spatial frequency conditions which preserves the subjective quality of the frames being masked. The quantitative model takes into account three fundamental properties of the video signals, namely the size of motion, average luminance and the power of each frequency components. The psychophysical responses towards the changes of these properties are obtained through subjective assessment tests using video sequences of simple geometrical patterns. Subjective experiments on natural video sequences show that more than 75% of viewers could make no distinction between the original sequence and the one processed using the quantitative model.

抄録全体を表示

A motion blur identification scheme is proposed for nonlinear uniform motion blurs approximated by piecewise linear models which consist of more than one linear motion component. The proposed scheme includes three modules that are a motion direction estimator, a motion length estimator and a motion combination selector. In order to identify the motion directions, the proposed scheme is based on a trial restoration by using directional forward ramp motion blurs along different directions and an analysis of directional information via frequency domain by using a Radon transform. Autocorrelation functions of image derivatives along several directions are employed for estimation of the motion lengths. A proper motion combination is identified by analyzing local autocorrelation functions of non-flat component of trial restored results. Experimental examples of simulated and real world blurred images are given to demonstrate a promising performance of the proposed scheme.

抄録全体を表示

In this study, we propose a complete architecture based on digital watermarking techniques to solve the issue of copyright protection and authentication for digital contents. We apply visible and semi-fragile watermarks as dual watermarks where visible watermarking is used to establish the copyright protection and semi-fragile watermarking authenticates and verifies the integrity of the watermarked image. In order to get the best tradeoff between the embedding energy of watermark and the perceptual translucence for visible watermark, the composite coefficients using global and local characteristics of the host and watermark images in the discrete wavelet transform (DWT) domain is considered with Human Vision System (HVS) models. To achieve the optimum noise reduction of the visibility thresholds for HVS in DWT domain, the contrast-sensitive function (CSF) and noise visible function (NVF) of perceptual model is applied which characterizes the global and local image properties and identifies texture and edge regions to determine the optimal watermark locations and strength at the watermark embedding stage. In addition, the perceptual weights according to the basis function amplitudes of DWT coefficients is fine tuned for the best quality of perceptual translucence in the design of the proposed watermarking algorithm. Furthermore, the semi-fragile watermark can detect and localize malicious attack effectively yet tolerate mild modifications such as JPEG compression and channel additive white Gaussian noise (AWGN). From the experimental results, our proposed technique not only improves the PSNR values and visual quality than other algorithms but also preserves the visibility of the watermark visible under various signal processing and advanced image recovery attacks.

抄録全体を表示

In this paper, we investigated the effect of chromaticity and luminance of surround to decide subject neutral white, and conducted a mathematical model of adapting degree for environment. Factors for adapting degree consist of two parts, adapting degree of ambient chromaticity and color saturation. These can be applied to color appearance models (CAM), actually improve the performance of color matching of CAM, hence would produce the method of image reproduction to general display systems.

抄録全体を表示

Digital video encapsulates the time series of a frame (still) images, where overall video quality can be obtained by using the quality of each frame image and the temporal information between the frame image. Coding of video produces degradation of these two types of information. These degradations can be classified as spatial degradation (static degradation) of a frame images and temporal degradation between frame image (dynamic degradation). In the framework of video quality evaluation it is necessary to consider those degradations, because their contents are strongly interdependable and quantification is problematic for these degradations. Therefore, the development of an objective video quality assessment method for single video quality requires to investigate how much static degradation and dynamic degradation affect single video quality. In this research, single video quality was predicted highly accuratly by using frame quality as static degradation and frame rate information as dynamic degradation.

抄録全体を表示

This paper presents a new method for reconstruction of trigonometric polynomials, a specific class of bandlimited signals, from a number of integrated values of input signals. It is applied in signal reconstruction, spectral estimation, system identification, as well as in other important signal processing problems. The proposed method of processing can be used for precise rms measurements of periodic signal (or power and energy) based on the presented signal reconstruction. Based on the value of the integral of the original input (analogue) signal, with a known frequency spectrum but unknown amplitudes and phases, a reconstruction of its basic parameters is done by the means of derived analytical and summarized expressions. Subsequent calculation of all relevant indicators related to the monitoring and processing of ac voltage and current signals is provided in this manner. Computer simulation demonstrating the precision of these algorithms. We investigate the errors related to the signal reconstruction, and provide an error bound around the reconstructed time domain waveform.

抄録全体を表示

This paper studies encoding/decoding function of artificial spiking neurons. First, we investigate basic characteristics of spike-trains of the neurons and fix parameter value that can minimize variation of spike-train length for initial value. Second we consider analog-to-digital encoding based upon spike-interval modulation that is suitable for simple and stable signal detection. Third we present a digital-to-analog decoder in which digital input is applied to switch the base signal of the spiking neuron. The system dynamics can be simplified into simple switched dynamical systems and precise analysis is possible. A simple circuit model is also presented.

抄録全体を表示

In the real world, it is not always true that neighboring houses are physically adjacent or close to each other. in other words, “neighbors” are not always “true neighbors” In this study, we propose a new Self-Organizing Map (SOM) algorithm, SOM with False-Neighbor degree between neurons (called FN-SOM). The behavior of FN-SOM is investigated with learning for various input data. We confirm that FN-SOM can obtain a more effective map reflecting the distribution state of input data than the conventional SOM and Growing Grid.

抄録全体を表示

In this paper, a generalized Montgomery multiplication algorithm in GF (2^m) using the Toeplitz matrix-vector representation is presented. The hardware architectures derived from this algorithm provide low-complexity bit-parallel systolic multipliers with trinomials and pentanomials. The results reveal that our proposed multipliers reduce the space complexity of approximately 15% compared with an existing systolic Montgomery multiplier for trinomials. Moreover, the proposed architectures have the features of regularity, modularity, and local interconnection. Accordingly, they are well suited to VLSI implementation.

抄録全体を表示

To improve the computation efficiency of the application specific instruction-set processor (ASIP), a strategy of hardware/software collaborative design is usually utilized. In this process, the autocustomization of specific instruction set has always been a key part to support the automated design of ASIP. The key issue of this problem is how to effectively reduce the huge exponential exploration space in the instruction identification process. To address this issue, we first formulate it as a feasible sub-graph enumeration problem under multiple constraints, and then propose a fast instruction identification algorithm based on a new model called basic convex pattern (BCP). The kernel technique in this algorithm is the transformation from the graph exploration to the formula-based computations. The experimental results have indicated that the proposed algorithm has a distinct reduction in the execution time.

抄録全体を表示

Research into applying LDPC code theory, which is used for channel coding, to source coding has received a lot of attention in several research fields such as distributed source coding. In this paper, a source coding problem with a fidelity criterion is considered. Matsunaga et al. and Martinian et al. constructed a lossy code under the conditions of a binary alphabet, a uniform distribution, and a Hamming measure of fidelity criterion. We extend their results and construct a lossy code under the extended conditions of a binary alphabet, a distribution that is not necessarily uniform, and a fidelity measure that is bounded and additive and show that the code can achieve the optimal rate, rate-distortion function. By applying a formula for the random walk on lattice to the analysis of LDPC matrices on Z_q, where q is a prime number, we show that results similar to those for the binary alphabet condition hold for Z_q, the multiple alphabet condition.

抄録全体を表示

We investigate the minimum weights of simple full-length array LDPC codes (SFA-LDPC codes). The SFA-LDPC codes are a subclass of LDPC codes, and constructed algebraically according to two integer parameters p and j. Mittelholzer and Yang et al. have studied the minimum weights of SFA-LDPC codes, but the exact minimum weights of the codes are not known except for some small p and j. In this paper, we show that the minimum weights of the SFA-LDPC codes with j=4 and j=5 are upper-bounded by 10 and 12, respectively, independent from the prime number p. By combining the results with Yang's lower-bound limits, we can conclude that the minimum weights of the SFA-LDPC codes with j=4 and p>7 are exactly 10 and those of the SFA-LDPC codes with j=5 are 10 or 12.

抄録全体を表示

This paper proposes a combination of adaptive equalizer and Least Mean Square-RAKE (LMS-RAKE) combining scheme receiver system for Direct Sequence-Ultra Wideband (DS-UWB) multipath channel model. The main purpose of the proposed system is to overcome the performance degradation for UWB transmission due to the occurrence of Inter-Symbol Interference (ISI) during high speed transmission of ultra short pulses in a multipath channel. The proposed system improves the system performance by mitigating the multipath effect using LMS-RAKE receiver and suppressing the ISI effect with the adaptive equalizer. Simulation results verify that significant equalization gain can be obtained by the proposed system especially in UWB multipath channel models such as channel CM3 and channel CM4 that suffered severe ISI effect.

抄録全体を表示

In this paper, a channel allocation scheme is studied for overlay wireless networks to optimize connection-level QoS. The contributions of our work are threefold. First, a channel allocation strategy using both horizontal channel borrowing and vertical traffic overflowing (HCBVTO) is presented and analyzed. When all the channels in a given macrocell are used, high-mobility real-time handoff requests can borrow channels from adjacent homogeneous cells. In case that the borrowing requests fail, handoff requests may also be overflowed to heterogeneous cells, if possible. Second, high-mobility real-time service is prioritized by allowing it to preempt channels currently used by other services. And third, to meet the high QoS requirements of some services and increase the utilization of radio resources, certain services can be transformed between real-time services and non-real-time services as necessary. Simulation results demonstrate that the proposed schemes can improve system performance.

抄録全体を表示

In order to reduce spatial redundancies, the H.264/AVC Intra coding provides nine directional prediction modes including DC prediction for every 4×4 block, but it needs a lot of overhead bits to represent the nine directional prediction modes for every 4×4 block. To compress the directional mode bits efficiently, the most probable mode is estimated by using the correlation between the prediction mode of spatially adjacent blocks and that of the current block. In this paper, a new method for estimating the most probable mode is proposed by using the directional information of the prediction mode of the adjacent blocks. Experimental results show that the proposed method is able to achieve a coding gain of about 0.2dB on average at low bit rate.

抄録全体を表示

A statistical model based on a partitioned Markov-chains model has previously been developed to represent time domain behavior of the asynchronous impulsive noise over a broadband power line communication (PLC) network. However, the estimation of its model parameters using the Simplex method can easily trap thee final solution at a local optimum. This study proposes an estimation scheme based on the genetic algorithm (GA) to overcome this difficulty. Experimental results show that the proposed scheme yields estimates that more closely match the experimental data statistics.

抄録全体を表示

A bit-depth scalability is proposed in an adaptive way based on modified inter-layer predictions of the spatial scalability. A simple prediction for high dynamic range (HDR) sequences is implemented to reduce the redundancy of the residual signals between the base layer which contains low dynamic range (LDR) sequences and the enhancement layer which contains HDR sequences by using scaling and offset values.

抄録全体を表示